Creating Multiple Cohorts Using the SAS DATA Step Jonathan Steinberg, Educational Testing Service, Princeton, NJ

Size: px
Start display at page:

Download "Creating Multiple Cohorts Using the SAS DATA Step Jonathan Steinberg, Educational Testing Service, Princeton, NJ"

Transcription

1 Creating Multiple Cohorts Using the SAS DATA Step Jonathan Steinberg, Educational Testing Service, Princeton, NJ ABSTRACT The challenge of creating multiple cohorts of people within a data set, based on one or more common characteristics, is driven by the nature of the data that is collected and how it is structured. A major benefit of establishing such a data set is the ease and flexibility with which you can statistically examine differences among respondents using various indicator variables, such as test dates, particularly if there is consistency to the data layout over time. If not, the data analyst may need to spend additional time to create such a layout. A standardized test may have several administrations during the course of a calendar year. The SAS frequency procedure, PROC FREQ, could easily identify patterns in the data. However, while some candidates will take a test once, others may take the same test multiple times. Each time a candidate takes a test, a record is appended to a database with the score, along with an appropriate indicator of whether or not the respondent is a repeat testtaker. An in-depth cohort analysis would focus on the patterns of testing dates, including test-retest behavior, as the time between testing dates may impact performance. There are two important issues to address at the outset. Multiple years of data may be needed to ensure adequate sample sizes for certain analyses. Additionally, the combination of administration date and repeat testtaking behavior can lead to a large series of frequency distributions that would need to be organized and interpreted, which may not be the most efficient way to start the cohort analysis. This paper will describe two DATA Step processes that are required before starting work on a project with the aim to improve the overall efficiency of the analysis. The first process is to create a single row for each respondent with the test score and repeater status for each possible administration. This requires combining several small files into one large file. The second process creates a single variable of repeater status across administrations. It is created within the new file so that testing dates, notably repeats, can be identified. This becomes the basis for the cohort analysis. This paper is intended for those with moderate experience in DATA Step processing. INTRODUCTION In general, when data is stored so that a date reference is a separate variable, there may be difficulties for the analyst if the scope of work involves restructuring the data to look at patterns of test-taking behavior based on time. The TABLES statement within PROC FREQ can easily generate counts of test-takers by a given date. However, that would not show patterns of multiple testing administrations. An issue that often surfaces when working with standardized test data is how to incorporate repeat test-taking behavior into the analysis. Repeater status is a variable often included in such data sets since most testing programs have many administrations during the course of a calendar year. Some candidates will take a test once while others will take the same test several times for various reasons, which sometimes can confound the results of the analysis. Thus, this requires more work than running a single PROC FREQ to examine the combination of testing dates and repeat test-taking behavior. The basis for this paper came from an attempt to explore patterns of repeat standardized test-taking over a rolling three-year period, identified by administration dates. The procedure by which the data set was rearranged will be described. The extraction of the test score information that follows is a relatively routine procedure using the DATA Step. Finally, it will be shown that more complex DATA Step processing is necessary to create timedependent respondent profiles of repeat test-taking status. The end result is a simplification of the research question upon which results from a single PROC FREQ can direct future analyses. 1

2 CREATION OF INDIVIDUAL TEST INFORMATION PROFILES Each time a respondent takes a test, a record is appended to the database with a unique identifier so that repeater status can easily be determined, along with possible other background information. The file is sorted by respondent identification number, calendar year, and month in ascending order. Table 1 displays the file layout along with simulated data. Table 1: Original File Layout with Simulated Data ID # Month Year Repeat Test Status Score Form N 179 A N 176 A N 180 A Y 163 A N 178 A N 161 B N 189 A N 182 A N 173 B N 157 A N 175 B N 187 A It is worth noting that the variable named Repeat Status has two possible values, N for first-time test-takers, and Y for repeat test-takers based on identification number. However, not all respondents may have first-time test-taking status in the data file. Given that this is a three-year snapshot into this particular testing environment, respondents may be classified as repeat test-takers because their first administration occurred before the beginning of the selected date range of the data file, as is the case in Row 4. Additionally, this test has multiple forms. A counter is applied beginning with the first instance of the respondent identifier after the data has been sorted by key variables. Table 2 displays an example of the resulting data set. proc sort data=test1 out=test1_sorted; by idnum year month; data test1_unique; set test1_sorted; by idnum year month; if first.idnum then records = 0; records + 1; 2

3 Table 2: Sorted Simulated Data File with Counter ID # Month Year Repeat Test Status Score Form Records N 160 A N 170 B N 184 B Y 134 A N 200 A Y 134 C Y 138 D Y 143 M N 186 C N 154 A Y 143 G Y 137 A Y 129 B Y 145 A Y 134 D 4 This extract from the data file shows that Respondent 6 took this test on three separate occasions. The four key variables extracted from the original database are the administration date (month and year), respondent identifier, repeater status, and test score, based on the last instance of the respondent identifier. This is accomplished using the KEEP statement on the DATA Step line: data test1_unique2 (keep = idnum month year repeat score); set test1_unique; by idnum; if last.idnum; The next step is to create individual data files by administration date of just the other three key variables mentioned above. For example: data test1_admin1102 (keep = idnum repeat score); set test1_sorted; where month = 11 and year = 2002; data test1_admin0103 (keep = idnum repeat score); set test1_sorted; where month = 1 and year = 2003; Then using the RENAME option, a suffix is appended to only the repeat status and score variables in the datespecific files, according to the particular administration date. Two examples are listed below: data test1_admin1102 (rename = (repeat = repeat1102 score = score1102)); set test1_admin1102; 3

4 data test1_admin0103 (rename = (repeat = repeat0103 score = score0103)); set test1_admin0103; Once completed for all administration dates, a data set consisting of only the identification number is created, again using the KEEP option. The smaller files are then merged together using the unique respondent identifier as the matching variable. data test1_unique3 (keep = idnum); set test1_unique2; data test1_total; merge test1_unique3 test1_admin1102 test1_admin0103; by idnum; Figure 3 displays a partial data layout after the merging process is completed: ID # Repeat Status Test Score (Nov 02) (Nov 02) N Repeat Status (Jan 03) Test Score (Jan 03) Y N Y N Y N N Y N Y 153 Y Y 145. It is evident from this small extract that the first four respondents did not take the test in November 2002 or in January This indicates that their first appearance in the data file occurred at a later administration date. Rows 6 through 12, as well as Row 16, have test data from November 2002 but not January 2003, with different values of repeat status. Rows 5, 13, and 14 have test data from January 2003, but not November 2002, again with varying values for repeat status. However, in Row 15, this respondent took this test both in November 2002 and January 2003 as a repeat test-taker, meaning his or her first testing date occurred prior to November IDENTIFICATION OF REPEAT TEST-TAKING BEHAVIOR PROFILES As displayed in Figure 3, the data is now arranged so that for each unique respondent in the original file, a set of key variable information grouped by administration date is displayed. As mentioned earlier, the key fields will be blank (for repeat test-taking status) or missing (for test score) for any administration date in which the respondent did not take the test. The creation of the repeater profile is done in two steps. First, given that the repeater status variable is text and not numeric, concatenation of these letters by administration is required. The width of the resulting variable, referred to as string, is equal to the number of administrations in the original data file. 4

5 data test1_total2; set test1_total; string = repeat1102 repeat0103 repeat0303 ; The second step is to use the FIND function to identify the column positions, expressed as numeric values from zero (indicating non-existence) to the number of administrations of four key test-taking dates, expressed as new variables: the instance of the N (if it exists), the instance of the first Y which marks the first repeat, the second Y, and the third Y, if any or all exist in the data file. firstn = find(string,'n',1); firsty = find(string,'y',1); secondy = find(string,'y',(firsty+1)); thirdy = find(string,'y',(secondy+1)); The function begins at the first column, and then for the variables called secondy and thirdy, the function begins from one column after finding the previous repeat. Please note that if an N exists for a respondent in the data file, the value of firsty will be greater than that of firstn, the value of secondy will be greater than that of firsty, and the value of thirdy will be greater than that of secondy. The resulting values for each of the four variables created above are contingent on the basis that they have a value other than zero in the data file. Figure 4 displays an example data set after this has been executed. It is worth noting the varying pattern of test-taking profiles, even among this subset of thirty observations. Most of these respondents only took the test once during the three-year period in the data file, but that first test date ranges from Date 2 to Date 17. Row 14 shows that the respondent took the test on three consecutive administration dates. Row 27 shows that at Time 16, the respondent took the test for the first time, and then repeated at Time 19. 5

6 Figure 4: Data File Extract after Creating Respondent Profiles A variable called status is appended to the data set above and is created based on the results from building the respondent profile. The goal here is to summarize the possible profiles in a way that will help inform the future cohort analysis and try to simplify the research question for the analyst. There are seven possible classifications formed in the following way: status =.; if firstn > 0 and firsty = 0 and secondy = 0 and thirdy = 0 then status = 1; if firstn = 0 and firsty > 0 and secondy = 0 then status = 2; if firstn > 0 and firsty > firstn and secondy = 0 then status = 3; if firstn = 0 and firsty > 0 and secondy > firsty and thirdy = 0 then status = 4; if firstn > 0 and firsty > firstn and secondy > firsty and thirdy = 0 then status = 5; 6

7 if firstn = 0 and firsty > 0 and secondy > firsty and thirdy > secondy then status = 6; if firstn > 0 and firsty > firstn and secondy > firsty and thirdy > secondy then status = 7; Table 5 summarizes the description of each group according to the status variable. The sample sizes generated from a frequency distribution of simulated data are used to examine repeat test-taking profile behavior patterns. Table 5: Summary of Respondent Profiles with Sample Proportions Status FirstN FirstY SecondY ThirdY Data Example % Sample 1 7,0,0, ,11,0, ,19,0, ,3,4, ,16,17, ,6,7, ,15,16, STUDY RESULTS Approximately 80% of this test-taking population took this test for the first and only time within the time frame of the database. That leaves the remaining 20% with varying degrees of repeat test-taking behavior. About threefourths of repeat test-takers (Groups 2, 4, and 6) will have both first taken the test prior to the start of the date range of the file, and repeated the test one, two, or three times, in approximately equal proportions. The remaining respondents took the test for the first time during the date range of the file, and then repeated mainly just once (about 2% as in Group 3), with others repeating more times (less than 1% each in Groups 5 and 7). IMPLICATIONS FOR FURTHER ANALYSES According to the results from this study, the next steps for Group 1 are the most straightforward as only a single test administration date defines the cohort, and is the easiest to explain since all of its members are first-time testtakers. The issue of test-taking behavior can now be explored for Group 1 through one PROC FREQ based on administration date. The sample size should be adequate for this group if the distribution of repeat test-taking status is similar to what was outlined in the study. However, sample size concerns may arise for the other groups, despite the apparent self-explanatory nature of respondent membership for Groups 3, 5, and 7. Groups 2, 4, and 6 would be affected if a need arises to define the first testing date, which would mean extending the date range of the file. The use of hierarchical clustering based on frequency distributions of, for example, values of firstn or firsty may be needed to alleviate problems of insufficient sample size. As mentioned previously, multiple years of data may be necessary for this type of analysis, but sufficient testing volume may also be a requirement to proceed, given the example cohort proportions reported here. The resulting data shown in this paper serves as the basis for asking questions such as: (1) Does there tend to be a clustering of respondents, whether first-time test-takers or repeaters, at certain administration dates? (2) What is the average change in test score if one repeats the test one, two, or three times? 7

8 (3) Given a three-year snapshot, does the effect on test performance if the periods between repeat administrations vary from a few months to as much as three years? CONCLUSION The procedure described in this paper takes into account a historically consistent data layout as applied to a standardized testing program. For the intended purpose of analyzing multiple cohorts of respondents based on testtaking behavior, considerations such as date range, the inclusion or exclusion of repeat test-takers, and perhaps testing volume are primary concerns to the data analyst before proceeding with a project such as this. A moderate level of DATA Step processing is required to carry out the tasks described in creating the cohorts for analysis, as the KEEP and RENAME options are used, as well as functions involving merging and finding text within a concatenated variable. The results shown through this analysis may or may not be typical of all testing programs, or all data types. However, the procedure is very powerful for easily creating multiple cohorts for further analyses, as well as identifying possible areas of research. This procedure aims to reduce the amount of output generated from what would otherwise be many uses of PROC FREQ and more importantly, can reduce the amount of additional exploration time expended by the data analyst. ACKNOWLEDGMENTS The author would like to thank Ted Blew, Catherine Trapani, and Jennifer Minsky for their support and assistance in proofreading this paper. REFERENCES SAS Online Help and Documentation, Cary, NC: SAS Institute Inc., Cody, Ronald P., and Smith, Jeffrey K. (1991) Applied Statistics and the SAS Programming Language, Third Edition. Englewood Cliffs, NJ: Prentice Hall. Feng, Ying (2006), PROC SQL: When and How to Use It?. Proceedings of 2006 NESUG Conference, Paper CC20. Online at Zhang, Rodger (2006), Creating a Report Showing Monthly, Quarter-To-Date and Year-To-Date Information Without Changing Date Parameters Monthly. Proceedings of 2006 NESUG Conference, Paper CC09. Online at TRADEMARKS SAS and all other SAS Institute, Inc. product or service names are registered trademarks or trademarks of SAS Institute, Inc. in the USA and other countries. indicates USA registration. Other brand and product names are registered trademarks or trademarks of their respective companies. 8

9 CONTACT INFORMATION Your comments and questions are appreciated and encouraged. Please contact the author at: Jonathan Steinberg Research Data Analyst Center for Data Analysis Research Educational Testing Service Rosedale Road Mail Stop 20-T Princeton, NJ Telephone: (609)

ABSTRACT INTRODUCTION

ABSTRACT INTRODUCTION Adaptive Randomization: Institutional Balancing Using SAS Macro Rita Tsang, Aptiv Solutions, Southborough, Massachusetts Katherine Kacena, Aptiv Solutions, Southborough, Massachusetts ABSTRACT Adaptive

More information

A SAS sy Study of ediary Data

A SAS sy Study of ediary Data A SAS sy Study of ediary Data ABSTRACT PharmaSUG 2017 - Paper BB14 A SAS sy Study of ediary Data Amie Bissonett, inventiv Health Clinical, Minneapolis, MN Many sponsors are using electronic diaries (ediaries)

More information

ABSTRACT THE INDEPENDENT MEANS T-TEST AND ALTERNATIVES SESUG Paper PO-10

ABSTRACT THE INDEPENDENT MEANS T-TEST AND ALTERNATIVES SESUG Paper PO-10 SESUG 01 Paper PO-10 PROC TTEST (Old Friend), What Are You Trying to Tell Us? Diep Nguyen, University of South Florida, Tampa, FL Patricia Rodríguez de Gil, University of South Florida, Tampa, FL Eun Sook

More information

Pharmaceutical Applications

Pharmaceutical Applications Creating an Input Dataset to Produce Incidence and Exposure Adjusted Special Interest AE's Pushpa Saranadasa, Merck & Co., Inc. Abstract Most adverse event (AE) summary tables display the number of patients

More information

PROGRAMMER S SAFETY KIT: Important Points to Remember While Programming or Validating Safety Tables

PROGRAMMER S SAFETY KIT: Important Points to Remember While Programming or Validating Safety Tables PharmaSUG 2012 - Paper IB09 PROGRAMMER S SAFETY KIT: Important Points to Remember While Programming or Validating Safety Tables Sneha Sarmukadam, Statistical Programmer, PharmaNet/i3, Pune, India Sandeep

More information

A SAS Macro to Present a Summary Table of the Number of Patients Having Experienced Adverse Events in a Clinical Trial

A SAS Macro to Present a Summary Table of the Number of Patients Having Experienced Adverse Events in a Clinical Trial A SAS Macro to Present a Summary Table of the Number of Patients Having Experienced Adverse Events in a Clinical Trial Christoph Gerlinger * and Ursula Franke ** * Laboratoires Fournier S.C.A. and ** biodat

More information

Generate Informative Clinical Laboratory Results Listing

Generate Informative Clinical Laboratory Results Listing PharmaSUG2011 - Paper PO02 Generate Informative Clinical Laboratory Results Listing Sai Ma, Everest Clinical Research Services Inc., Markham, Ontario Canada ABSTRACT Clinical laboratory results review

More information

Parameter Estimation of Cognitive Attributes using the Crossed Random- Effects Linear Logistic Test Model with PROC GLIMMIX

Parameter Estimation of Cognitive Attributes using the Crossed Random- Effects Linear Logistic Test Model with PROC GLIMMIX Paper 1766-2014 Parameter Estimation of Cognitive Attributes using the Crossed Random- Effects Linear Logistic Test Model with PROC GLIMMIX ABSTRACT Chunhua Cao, Yan Wang, Yi-Hsin Chen, Isaac Y. Li University

More information

Chapter 3 Software Packages to Install How to Set Up Python Eclipse How to Set Up Eclipse... 42

Chapter 3 Software Packages to Install How to Set Up Python Eclipse How to Set Up Eclipse... 42 Table of Contents Preface..... 21 About the Authors... 23 Acknowledgments... 24 How This Book is Organized... 24 Who Should Buy This Book?... 24 Where to Find Answers to Review Questions and Exercises...

More information

Programmatic Challenges of Dose Tapering Using SAS

Programmatic Challenges of Dose Tapering Using SAS ABSTRACT: Paper 2036-2014 Programmatic Challenges of Dose Tapering Using SAS Iuliana Barbalau, Santen Inc., Emeryville, CA Chen Shi, Santen Inc., Emeryville, CA Yang Yang, Santen Inc., Emeryville, CA In

More information

Paper number: CC02 MACRO %NEWFLOW JIAN HUA (DANIEL) HUANG, FOREST LABORATORIES INC, NJ

Paper number: CC02 MACRO %NEWFLOW JIAN HUA (DANIEL) HUANG, FOREST LABORATORIES INC, NJ Paper number: CC02 MACRO %NEWFLOW JIAN HUA (DANIEL) HUANG, FOREST LABORATORIES INC, NJ ABSTRACT: If you use SAS Proc Report frequently, you will be familiar with the flow option. The flow option wraps

More information

Review Questions in Introductory Knowledge... 37

Review Questions in Introductory Knowledge... 37 Table of Contents Preface..... 17 About the Authors... 19 How This Book is Organized... 20 Who Should Buy This Book?... 20 Where to Find Answers to Review Questions and Exercises... 20 How to Report Errata...

More information

Extract Information from Large Database Using SAS Array, PROC FREQ, and SAS Macro

Extract Information from Large Database Using SAS Array, PROC FREQ, and SAS Macro SESUG 2016 Paper CC-160 Extract Information from Large Database Using SAS Array, PROC FREQ, and SAS Macro Lifang Zhang, Department of Biostatistics and Epidemiology, Augusta University ABSTRACT SAS software

More information

Mass Modification User Guide for Service Providers and Service Provider Consultants

Mass Modification User Guide for Service Providers and Service Provider Consultants Mass Modification User Guide for Service Providers and Service Provider Consultants Version 3.0 January 25, 2018 Change History DATE REVISION DESCRIPTION 03/19/2012 1.0 New Guide 05/31/2017 2.0 Update

More information

Combining Electronic Diary Seizure Data with Visit-Based Seizure Data in the MONEAD Study

Combining Electronic Diary Seizure Data with Visit-Based Seizure Data in the MONEAD Study PharmaSUG 2018 - Paper RW-02 Combining Electronic Diary Seizure Data with Visit-Based Seizure Data in the MONEAD Study Julia Skinner and Ryan May, Emmes Corporation ABSTRACT Electronic diary (ediary) mobile

More information

Registered Radiologist Assistant (R.R.A. ) 2016 Examination Statistics

Registered Radiologist Assistant (R.R.A. ) 2016 Examination Statistics Registered Radiologist Assistant (R.R.A. ) Examination Statistics INTRODUCTION This report summarizes the results of the Registered Radiologist Assistant (R.R.A. ) examinations developed and administered

More information

Chapter 1: Managing workbooks

Chapter 1: Managing workbooks Chapter 1: Managing workbooks Module A: Managing worksheets You use the Insert tab on the ribbon to insert new worksheets. True or False? Which of the following are options for moving or copying a worksheet?

More information

The Creative Porpoise Revisited

The Creative Porpoise Revisited EUROPEAN JOURNAL OF BEHAVIOR ANALYSIS 2012, 13, xx - xx NUMBER 1 (SUMMER 2012) 1 The Creative Porpoise Revisited Per Holth Oslo and Akershus University College The sources of novel behavior and behavioral

More information

Quasicomplete Separation in Logistic Regression: A Medical Example

Quasicomplete Separation in Logistic Regression: A Medical Example Quasicomplete Separation in Logistic Regression: A Medical Example Madeline J Boyle, Carolinas Medical Center, Charlotte, NC ABSTRACT Logistic regression can be used to model the relationship between a

More information

ADaM Tips for Exposure Summary

ADaM Tips for Exposure Summary ABSTRACT PharmaSUG China 2016 Paper 84 ADaM Tips for Exposure Summary Kriss Harris, SAS Specialists Limited, Hertfordshire, United Kingdom Have you ever created the Analysis Dataset for the Exposure Summary

More information

Quality Improvement of Causes of Death Statistics by Automated Coding in Estonia, 2011

Quality Improvement of Causes of Death Statistics by Automated Coding in Estonia, 2011 Quality Improvement of Causes of Death Statistics by Automated Coding in Estonia, 2011 Technical Implementation report, Grant agreement nr 10501.2009.002-2009.461 Introduction The grant agreement between

More information

Collapsing Longitudinal Data Across Related Events and Imputing Endpoints

Collapsing Longitudinal Data Across Related Events and Imputing Endpoints Collapsing Longitudinal Data Across Related Events and Imputing Endpoints James Joseph, INC Research, King of Prussia, PA INTRODUCTION When attempting to answer a research question, the best study design

More information

Imputing Dose Levels for Adverse Events

Imputing Dose Levels for Adverse Events Paper HO03 Imputing Dose Levels for Adverse Events John R Gerlach & Igor Kolodezh CSG Inc., Raleigh, NC USA ABSTRACT Besides the standard reporting of adverse events in clinical trials, there is a growing

More information

About Reading Scientific Studies

About Reading Scientific Studies About Reading Scientific Studies TABLE OF CONTENTS About Reading Scientific Studies... 1 Why are these skills important?... 1 Create a Checklist... 1 Introduction... 1 Abstract... 1 Background... 2 Methods...

More information

Reveal Relationships in Categorical Data

Reveal Relationships in Categorical Data SPSS Categories 15.0 Specifications Reveal Relationships in Categorical Data Unleash the full potential of your data through perceptual mapping, optimal scaling, preference scaling, and dimension reduction

More information

OneTouch Reveal Web Application. User Manual for Healthcare Professionals Instructions for Use

OneTouch Reveal Web Application. User Manual for Healthcare Professionals Instructions for Use OneTouch Reveal Web Application User Manual for Healthcare Professionals Instructions for Use Contents 2 Contents Chapter 1: Introduction...4 Product Overview...4 Intended Use...4 System Requirements...

More information

Knowledge is Power: The Basics of SAS Proc Power

Knowledge is Power: The Basics of SAS Proc Power ABSTRACT Knowledge is Power: The Basics of SAS Proc Power Elaina Gates, California Polytechnic State University, San Luis Obispo There are many statistics applications where it is important to understand

More information

Blast Searcher Formative Evaluation. March 02, Adam Klinger and Josh Gutwill

Blast Searcher Formative Evaluation. March 02, Adam Klinger and Josh Gutwill Blast Searcher Formative Evaluation March 02, 2006 Keywords: Genentech Biology Life Sciences Adam Klinger and Josh Gutwill Traits of Life \ - 1 - BLAST Searcher Formative Evaluation March 2, 2006 Adam

More information

Managing Immunizations

Managing Immunizations Managing Immunizations In this chapter: Viewing Immunization Information Entering Immunizations Editing Immunizations Entering a Lead Test Action Editing a Lead Test Action Entering Opt-Out Immunizations

More information

TABLE OF CONTENTS. Introduction

TABLE OF CONTENTS. Introduction TABLE OF CONTENTS Introduction Level 5 exemplar 1 Level 5 exemplar 2 Level 4 exemplar 1 Level 4 exemplar 2 Level 3 exemplar 1 Level 3 exemplar 2 Level 2 exemplar 1 Level 2 exemplar 2 Level 1 exemplar 1

More information

Spatial Orientation Using Map Displays: A Model of the Influence of Target Location

Spatial Orientation Using Map Displays: A Model of the Influence of Target Location Gunzelmann, G., & Anderson, J. R. (2004). Spatial orientation using map displays: A model of the influence of target location. In K. Forbus, D. Gentner, and T. Regier (Eds.), Proceedings of the Twenty-Sixth

More information

Methodology for Non-Randomized Clinical Trials: Propensity Score Analysis Dan Conroy, Ph.D., inventiv Health, Burlington, MA

Methodology for Non-Randomized Clinical Trials: Propensity Score Analysis Dan Conroy, Ph.D., inventiv Health, Burlington, MA PharmaSUG 2014 - Paper SP08 Methodology for Non-Randomized Clinical Trials: Propensity Score Analysis Dan Conroy, Ph.D., inventiv Health, Burlington, MA ABSTRACT Randomized clinical trials serve as the

More information

Clay Tablet Connector for hybris. User Guide. Version 1.5.0

Clay Tablet Connector for hybris. User Guide. Version 1.5.0 Clay Tablet Connector for hybris User Guide Version 1.5.0 August 4, 2016 Copyright Copyright 2005-2016 Clay Tablet Technologies Inc. All rights reserved. All rights reserved. This document and its content

More information

SUPPLEMENTARY APPENDICES FOR ONLINE PUBLICATION. Supplement to: All-Cause Mortality Reductions from. Measles Catch-Up Campaigns in Africa

SUPPLEMENTARY APPENDICES FOR ONLINE PUBLICATION. Supplement to: All-Cause Mortality Reductions from. Measles Catch-Up Campaigns in Africa SUPPLEMENTARY APPENDICES FOR ONLINE PUBLICATION Supplement to: All-Cause Mortality Reductions from Measles Catch-Up Campaigns in Africa (by Ariel BenYishay and Keith Kranker) 1 APPENDIX A: DATA DHS survey

More information

Chapter 1. Introduction

Chapter 1. Introduction Chapter 1 Introduction 1.1 Motivation and Goals The increasing availability and decreasing cost of high-throughput (HT) technologies coupled with the availability of computational tools and data form a

More information

Stata: Merge and append Topics: Merging datasets, appending datasets - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 1. Terms There are several situations when working with

More information

Using SAS to Calculate Tests of Cliff s Delta. Kristine Y. Hogarty and Jeffrey D. Kromrey

Using SAS to Calculate Tests of Cliff s Delta. Kristine Y. Hogarty and Jeffrey D. Kromrey Using SAS to Calculate Tests of Cliff s Delta Kristine Y. Hogarty and Jeffrey D. Kromrey Department of Educational Measurement and Research, University of South Florida ABSTRACT This paper discusses a

More information

A SAS Macro to Investigate Statistical Power in Meta-analysis Jin Liu, Fan Pan University of South Carolina Columbia

A SAS Macro to Investigate Statistical Power in Meta-analysis Jin Liu, Fan Pan University of South Carolina Columbia Paper 109 A SAS Macro to Investigate Statistical Power in Meta-analysis Jin Liu, Fan Pan University of South Carolina Columbia ABSTRACT Meta-analysis is a quantitative review method, which synthesizes

More information

FAQs about Provider Profiles on Breast Cancer Screenings (Mammography) Q: Who receives a profile on breast cancer screenings (mammograms)?

FAQs about Provider Profiles on Breast Cancer Screenings (Mammography) Q: Who receives a profile on breast cancer screenings (mammograms)? FAQs about Provider Profiles on Breast Cancer Screenings (Mammography) Q: Who receives a profile on breast cancer screenings (mammograms)? A: We send letters and/or profiles to PCPs with female members

More information

Hearing aid dispenser approval process review Introduction Hearing aid dispenser data transfer... 6

Hearing aid dispenser approval process review Introduction Hearing aid dispenser data transfer... 6 Hearing aid dispenser approval process review 2010 11 Content 1.0 Introduction... 4 1.1 About this document... 4 1.2 Overview of the approval process... 4 2.0 Hearing aid dispenser data transfer... 6 2.1

More information

HOMEWORK 4 Due: next class 2/8

HOMEWORK 4 Due: next class 2/8 HOMEWORK 4 Due: next class 2/8 1. Recall the class data we collected concerning body image (about right, overweight, underweight). Following the body image example in OLI, answer the following question

More information

Case study examining the impact of German reunification on life expectancy

Case study examining the impact of German reunification on life expectancy Supplementary Materials 2 Case study examining the impact of German reunification on life expectancy Table A1 summarises our case study. This is a simplified analysis for illustration only and does not

More information

Using Lertap 5 in a Parallel-Forms Reliability Study

Using Lertap 5 in a Parallel-Forms Reliability Study Lertap 5 documents series. Using Lertap 5 in a Parallel-Forms Reliability Study Larry R Nelson Last updated: 16 July 2003. (Click here to branch to www.lertap.curtin.edu.au.) This page has been published

More information

Technologies for Data Analysis for Experimental Biologists

Technologies for Data Analysis for Experimental Biologists Technologies for Data Analysis for Experimental Biologists Melanie I Stefan, PhD HMS 15 May 2014 MI Stefan (HMS) Data Analysis with JMP 15 May 2014 1 / 35 Before we start Download files for today s class

More information

Implementing Worst Rank Imputation Using SAS

Implementing Worst Rank Imputation Using SAS Paper SP12 Implementing Worst Rank Imputation Using SAS Qian Wang, Merck Sharp & Dohme (Europe), Inc., Brussels, Belgium Eric Qi, Merck & Company, Inc., Upper Gwynedd, PA ABSTRACT Classic designs of randomized

More information

Author's response to reviews

Author's response to reviews Author's response to reviews Title: Morbidity and doctor characteristics only partly explain the substantial healthcare expenditures of frequent attenders - A record linkage study between patient data

More information

PROC SQL. By Becky Leung Alberta Health Services. Calgary SAS User Group Meeting Wednesday, October 08, 2014

PROC SQL. By Becky Leung Alberta Health Services. Calgary SAS User Group Meeting Wednesday, October 08, 2014 PROC SQL By Becky Leung Alberta Health Services Calgary SAS User Group Meeting Wednesday, October 08, 2014 WHAT IS PROC SQL? A Base SAS procedure that combines the functionality of DATA and PROC steps

More information

Web Based Instruction on Solar Eclipse and Lunar Eclipse for Deaf and Hard of Hearing People

Web Based Instruction on Solar Eclipse and Lunar Eclipse for Deaf and Hard of Hearing People Web Based Instruction on Solar Eclipse and Lunar Eclipse for Deaf and Hard of Hearing People Abstract Benjaporn Saksiri 1 Paruhut Suphajanya 2 Ratchasuda College, Madidol University, Thailand rsbss@mahidol.ac.th

More information

Adaptive Aspirations in an American Financial Services Organization: A Field Study

Adaptive Aspirations in an American Financial Services Organization: A Field Study Adaptive Aspirations in an American Financial Services Organization: A Field Study Stephen J. Mezias Department of Management and Organizational Behavior Leonard N. Stern School of Business New York University

More information

USE AND MISUSE OF MIXED MODEL ANALYSIS VARIANCE IN ECOLOGICAL STUDIES1

USE AND MISUSE OF MIXED MODEL ANALYSIS VARIANCE IN ECOLOGICAL STUDIES1 Ecology, 75(3), 1994, pp. 717-722 c) 1994 by the Ecological Society of America USE AND MISUSE OF MIXED MODEL ANALYSIS VARIANCE IN ECOLOGICAL STUDIES1 OF CYNTHIA C. BENNINGTON Department of Biology, West

More information

International Pharmaceutical Aerosol Consortium on Regulation and Science

International Pharmaceutical Aerosol Consortium on Regulation and Science International Pharmaceutical Aerosol Consortium on Regulation and Science 1500 K Street NW Washington DC 20005 Telephone +1 202 230 5607 Fax +1 202 842 8465 Email info@ipacrs.org Web www.ipacrs.org Submitted

More information

Inverse Probability of Censoring Weighting for Selective Crossover in Oncology Clinical Trials.

Inverse Probability of Censoring Weighting for Selective Crossover in Oncology Clinical Trials. Paper SP02 Inverse Probability of Censoring Weighting for Selective Crossover in Oncology Clinical Trials. José Luis Jiménez-Moro (PharmaMar, Madrid, Spain) Javier Gómez (PharmaMar, Madrid, Spain) ABSTRACT

More information

Ref: E 007. PGEU Response. Consultation on measures for improving the recognition of medical prescriptions issued in another Member State

Ref: E 007. PGEU Response. Consultation on measures for improving the recognition of medical prescriptions issued in another Member State Ref:11.11.24E 007 PGEU Response Consultation on measures for improving the recognition of medical prescriptions issued in another Member State PGEU The Pharmaceutical Group of the European Union (PGEU)

More information

Spending estimates from Cancer Care Spending

Spending estimates from Cancer Care Spending CALIFORNIA HEALTHCARE FOUNDATION August 2015 Estimating Cancer Care Spending in the California Medicare Population: Methodology Detail This paper describes in detail the methods used by Deborah Schrag,

More information

SAMPLING ERROI~ IN THE INTEGRATED sysrem FOR SURVEY ANALYSIS (ISSA)

SAMPLING ERROI~ IN THE INTEGRATED sysrem FOR SURVEY ANALYSIS (ISSA) SAMPLING ERROI~ IN THE INTEGRATED sysrem FOR SURVEY ANALYSIS (ISSA) Guillermo Rojas, Alfredo Aliaga, Macro International 8850 Stanford Boulevard Suite 4000, Columbia, MD 21045 I-INTRODUCTION. This paper

More information

Systematic reviews and meta-analyses of observational studies (MOOSE): Checklist.

Systematic reviews and meta-analyses of observational studies (MOOSE): Checklist. Systematic reviews and meta-analyses of observational studies (MOOSE): Checklist. MOOSE Checklist Infliximab reduces hospitalizations and surgery interventions in patients with inflammatory bowel disease:

More information

Transplant Recipients International Organization, Inc. Strategic Plan. Board Approved September 28, 2012

Transplant Recipients International Organization, Inc. Strategic Plan. Board Approved September 28, 2012 Transplant Recipients International Organization, Inc. Strategic Plan Board Approved September 28, 2012 TRIO Mission TRIO is a non-profit international organization committed to improving the quality of

More information

Education and Training Committee 15 November 2012

Education and Training Committee 15 November 2012 Education and Training Committee 15 November 2012 Review of the process of approval of hearing aid dispenser pre-registration education and training programmes. Executive summary and recommendations Introduction

More information

GENERAL INFORMATION AND INSTRUCTIONS

GENERAL INFORMATION AND INSTRUCTIONS NON-PARTICIPATING MANUFACTURER CERTIFICATION FOR LISTING ON OREGON DIRECTORY GENERAL INFORMATION AND INSTRUCTIONS Who is required to file this Certification? Any tobacco product manufacturer who is a non-participating

More information

M A N I T O B A ) Order No. 121/12 ) THE PUBLIC UTILITIES BOARD ACT ) September 12, 2012

M A N I T O B A ) Order No. 121/12 ) THE PUBLIC UTILITIES BOARD ACT ) September 12, 2012 M A N I T O B A ) ) THE PUBLIC UTILITIES BOARD ACT ) September 12, 2012 BEFORE: Susan Proven, P.H.Ec., Acting Chair Robert Warren, MBA, Member Regis Gosselin, BA, MBA, CGA, Chair TOWN OF LAC DU BONNET

More information

Kansas Bureau of Investigation

Kansas Bureau of Investigation Kirk Thompson Director Kansas Bureau of Investigation EXECUTIVE SUMMARY -7- The Kansas Sexual Assault Kit Initiative (SAKI): Future Sexual Assault Kit Submission and Testing May 10, 2018 Derek Schmidt

More information

Minnesota State Board of Assessors

Minnesota State Board of Assessors Minnesota State Board of Assessors DUAL NOTICE: Notice of Intent to Adopt Rules Without a Public Hearing Unless 25 or More Persons Request a Hearing, and Notice of Hearing if 25 or More Requests for Hearing

More information

How Many Options do Multiple-Choice Questions Really Have?

How Many Options do Multiple-Choice Questions Really Have? How Many Options do Multiple-Choice Questions Really Have? ABSTRACT One of the major difficulties perhaps the major difficulty in composing multiple-choice questions is the writing of distractors, i.e.,

More information

National Deaf Children s Society Response to Committee Report Education and Culture Committee Inquiry: attainment of pupils with a sensory impairment

National Deaf Children s Society Response to Committee Report Education and Culture Committee Inquiry: attainment of pupils with a sensory impairment National Deaf Children s Society Response to Committee Report Education and Culture Committee Inquiry: attainment of pupils with a sensory impairment 7 October 2015 1. Context We welcome the Committee

More information

Data Management System (DMS) User Guide

Data Management System (DMS) User Guide Data Management System (DMS) User Guide Eversense and the Eversense logo are trademarks of Senseonics, Incorporated. Other brands and their products are trademarks or registered trademarks of their respective

More information

Part IV: Interim Assessment Hand Scoring System

Part IV: Interim Assessment Hand Scoring System Interim Assessment Hand Scoring System Overview of the Overview of the Interim Assessment Hand Scoring System The (IAHSS) allows educators to score responses to items that require hand scoring. When students

More information

A Strategy for Handling Missing Data in the Longitudinal Study of Young People in England (LSYPE)

A Strategy for Handling Missing Data in the Longitudinal Study of Young People in England (LSYPE) Research Report DCSF-RW086 A Strategy for Handling Missing Data in the Longitudinal Study of Young People in England (LSYPE) Andrea Piesse and Graham Kalton Westat Research Report No DCSF-RW086 A Strategy

More information

Functional Assessment Observation Form

Functional Assessment Observation Form Functional Assessment Observation Form THE CONTENT OF THE FUNCTIONAL ASSESSMENT OBSERVATION FORM This Functional Assessment Observation Form has eight major sections (see next page). A blank copy of the

More information

AE: An Essential Part of Safety Summary Table Creation. Rucha Landge, Inventiv International Pharma Services Pvt. Ltd.

AE: An Essential Part of Safety Summary Table Creation. Rucha Landge, Inventiv International Pharma Services Pvt. Ltd. PharmaSUG 2016 - Paper IB11 AE: An Essential Part of Safety Summary Table Creation. Rucha Landge, Inventiv International Pharma Services Pvt. Ltd., Pune, India ABSTRACT Adverse event (AE) summary tables

More information

Does factor indeterminacy matter in multi-dimensional item response theory?

Does factor indeterminacy matter in multi-dimensional item response theory? ABSTRACT Paper 957-2017 Does factor indeterminacy matter in multi-dimensional item response theory? Chong Ho Yu, Ph.D., Azusa Pacific University This paper aims to illustrate proper applications of multi-dimensional

More information

Diffusion of Computer Applications Among Physicians: A Quasi-Experimental Study

Diffusion of Computer Applications Among Physicians: A Quasi-Experimental Study Clinical Sociology Review Volume 8 Issue 1 Article 10 1-1-1990 Diffusion of Computer Applications Among Physicians: A Quasi-Experimental Study James G. Anderson Purdue University Stephen J. Jay Methodist

More information

Judy Li Nick Chen The Quit Group

Judy Li Nick Chen The Quit Group Redemption of Nicotine Replacement Therapy (NRT) Quit Cards distributed through the Quitline, January June 2007 Judy Li Nick Chen The Quit Group July 2008 1 EXECUTIVE SUMMARY Aims 1. To give an indication

More information

Negative Effects of Using List Items as Recall Cues 1

Negative Effects of Using List Items as Recall Cues 1 JOURNAL OF VERBAL LEARNING AND VERBAL BEHAVIOR 12, 43-50 (1973) Negative Effects of Using List Items as Recall Cues 1 DEWEY RUNDUS 2 The Rockefeller University, New York, New York 10021 It is proposed

More information

BIOL 458 BIOMETRY Lab 7 Multi-Factor ANOVA

BIOL 458 BIOMETRY Lab 7 Multi-Factor ANOVA BIOL 458 BIOMETRY Lab 7 Multi-Factor ANOVA PART 1: Introduction to Factorial ANOVA ingle factor or One - Way Analysis of Variance can be used to test the null hypothesis that k or more treatment or group

More information

Introduction to Survival Analysis Procedures (Chapter)

Introduction to Survival Analysis Procedures (Chapter) SAS/STAT 9.3 User s Guide Introduction to Survival Analysis Procedures (Chapter) SAS Documentation This document is an individual chapter from SAS/STAT 9.3 User s Guide. The correct bibliographic citation

More information

EPO-144 Patients with Morbid Obesity and Congestive Heart Failure Have Longer Operative Time and Room Time in Total Hip Arthroplasty

EPO-144 Patients with Morbid Obesity and Congestive Heart Failure Have Longer Operative Time and Room Time in Total Hip Arthroplasty SESUG 2016 EPO-144 Patients with Morbid Obesity and Congestive Heart Failure Have Longer Operative Time and Room Time in Total Hip Arthroplasty ABSTRACT Yubo Gao, University of Iowa Hospitals and Clinics,

More information

Manitoba Annual Immunization Surveillance Report

Manitoba Annual Immunization Surveillance Report Annual Immunization Surveillance Report January 1 to December 31, 2014 Epidemiology & Surveillance Public Branch Public and Primary Care Division, y Living and Seniors Released: January 2016 TABLE OF CONTENTS

More information

INSTRUCTIONS FOR AUTHORS

INSTRUCTIONS FOR AUTHORS INSTRUCTIONS FOR AUTHORS Chinese Journal of Integrative Medicine is a peer-reviewed monthly journal sponsored by Chinese Association of Integrative Medicine and China Academy of Chinese Medical Sciences.

More information

APPENDIX B TAP 31 RESOLUTION PROCESS

APPENDIX B TAP 31 RESOLUTION PROCESS I. General Principles APPENDIX B TAP 31 RESOLUTION PROCESS a. Administration: The TAP 31 Resolution Process is administered by the University s Title IX Coordinator and the University s Deputy Title IX

More information

Drug Surveillance and Effectiveness Epidemiology Research - Questions to Ask your Epidemiologist

Drug Surveillance and Effectiveness Epidemiology Research - Questions to Ask your Epidemiologist Drug Surveillance and Effectiveness Epidemiology Research - Questions to Ask your Epidemiologist Roberta Glass, OptumInsight Life Sciences, Waltham, Massachusetts ABSTRACT As a SAS analyst starting work

More information

JSM Survey Research Methods Section

JSM Survey Research Methods Section Methods and Issues in Trimming Extreme Weights in Sample Surveys Frank Potter and Yuhong Zheng Mathematica Policy Research, P.O. Box 393, Princeton, NJ 08543 Abstract In survey sampling practice, unequal

More information

NARxCHECK Score as a Predictor of Unintentional Overdose Death

NARxCHECK Score as a Predictor of Unintentional Overdose Death NARxCHECK Score as a Predictor of Unintentional Overdose Death Huizenga J.E., Breneman B.C., Patel V.R., Raz A., Speights D.B. October 2016 Appriss, Inc. NOTE: This paper was previously published with

More information

Lesson: A Ten Minute Course in Epidemiology

Lesson: A Ten Minute Course in Epidemiology Lesson: A Ten Minute Course in Epidemiology This lesson investigates whether childhood circumcision reduces the risk of acquiring genital herpes in men. 1. To open the data we click on File>Example Data

More information

Reshaping of Human Fertility Database data from long to wide format in Excel

Reshaping of Human Fertility Database data from long to wide format in Excel Max-Planck-Institut für demografische Forschung Max Planck Institute for Demographic Research Konrad-Zuse-Strasse 1 D-18057 Rostock GERMANY Tel +49 (0) 3 81 20 81-0; Fax +49 (0) 3 81 20 81-202; http://www.demogr.mpg.de

More information

Technical Assistance Guide No Recommended PDMP Reports to Support Licensing/Regulatory Boards and Law Enforcement Investigations

Technical Assistance Guide No Recommended PDMP Reports to Support Licensing/Regulatory Boards and Law Enforcement Investigations Technical Assistance Guide No. 02-14 Recommended PDMP Reports to Support Licensing/Regulatory Boards and Law Enforcement Investigations This project was supported by Grant No. 2011-PM-BX-K001 awarded by

More information

NAME OF HEALTH UNIT NAME OF SUB-DISTRICT NAME OF DISTRICT

NAME OF HEALTH UNIT NAME OF SUB-DISTRICT NAME OF DISTRICT THE REPUBLIC OF UGANDA NATIONAL HIV CARE QUARTERY MONITORING REPORTING FORM MINISTRY OF HEALTH NAME OF HEALTH UNIT NAME OF SUB-DISTRICT NAME OF DISTRICT START END Instruction for tabulating the HIV care/

More information

Dementia Direct Enhanced Service

Dementia Direct Enhanced Service Vision 3 Dementia Direct Enhanced Service England Outcomes Manager Copyright INPS Ltd 2015 The Bread Factory, 1A Broughton Street, Battersea, London, SW8 3QJ T: +44 (0) 207 501700 F:+44 (0) 207 5017100

More information

Transitions in Depressive Symptoms After 10 Years of Follow-up Using PROC LTA

Transitions in Depressive Symptoms After 10 Years of Follow-up Using PROC LTA PharmaSUG 2015 Paper QT25 Transitions in Depressive Symptoms After 10 Years of Follow-up Using PROC LTA Seungyoung Hwang, Johns Hopkins University Bloomberg School of Public Health ABSTRACT PROC LTA is

More information

THE AFIX PRODUCT TRAINING MANUAL

THE AFIX PRODUCT TRAINING MANUAL THE AFIX PRODUCT TRAINING MANUAL Last Updated: 11/30/2018 Table of Contents The AFIX Product End User Training AFIX Cohort. 4 Provider Selection... 4 Assessment Selection..... 7 Reports.......10 Flexible

More information

Modeling Sentiment with Ridge Regression

Modeling Sentiment with Ridge Regression Modeling Sentiment with Ridge Regression Luke Segars 2/20/2012 The goal of this project was to generate a linear sentiment model for classifying Amazon book reviews according to their star rank. More generally,

More information

Pharmaceutical Trade Marks

Pharmaceutical Trade Marks 1 Pharmaceutical Trade Marks September 2015 Ben Mooneapillay bmooneapillay@jakemp.com www.jakemp.com C 22 H 30 N 6 O 4 S C 26 H 44 N 2 O 10 S sildenafil citrate salbutamol sulphate Badge of Origin Quality

More information

CARE Cross-project Collectives Analysis: Technical Appendix

CARE Cross-project Collectives Analysis: Technical Appendix CARE Cross-project Collectives Analysis: Technical Appendix The CARE approach to development support and women s economic empowerment is often based on the building blocks of collectives. Some of these

More information

Chapter 1: Data Collection Pearson Prentice Hall. All rights reserved

Chapter 1: Data Collection Pearson Prentice Hall. All rights reserved Chapter 1: Data Collection 2010 Pearson Prentice Hall. All rights reserved 1-1 Statistics is the science of collecting, organizing, summarizing, and analyzing information to draw conclusions or answer

More information

Using Vision Searches and Reports Module Tools for Disease Register Maintenance in a Post QOF World

Using Vision Searches and Reports Module Tools for Disease Register Maintenance in a Post QOF World Using Vision Searches and Reports Module Tools for Disease Register Maintenance in a Post QOF World Andrew Vickerstaff SCIMP panel member Practice Manager, Aultbea & Gairloch Medical Practice, NHS Highland

More information

Non Linear Control of Glycaemia in Type 1 Diabetic Patients

Non Linear Control of Glycaemia in Type 1 Diabetic Patients Non Linear Control of Glycaemia in Type 1 Diabetic Patients Mosè Galluzzo*, Bartolomeo Cosenza Dipartimento di Ingegneria Chimica dei Processi e dei Materiali, Università degli Studi di Palermo Viale delle

More information

Chapter 2 A Guide to Implementing Quantitative Bias Analysis

Chapter 2 A Guide to Implementing Quantitative Bias Analysis Chapter 2 A Guide to Implementing Quantitative Bias Analysis Introduction Estimates of association from nonrandomized epidemiologic studies are susceptible to two types of error: random error and systematic

More information

Evolutionary Programming

Evolutionary Programming Evolutionary Programming Searching Problem Spaces William Power April 24, 2016 1 Evolutionary Programming Can we solve problems by mi:micing the evolutionary process? Evolutionary programming is a methodology

More information

Associate Board Description Sheet

Associate Board Description Sheet The Night Ministry (TNM) is a Chicago-based organization that works to provide housing, health care and human connection to members of our community struggling with poverty or homelessness. The Night Ministry

More information

Re: Bill S-5, An Act to amend the Tobacco Act and the Non-smokers Health Act and to make consequential amendments to other Acts

Re: Bill S-5, An Act to amend the Tobacco Act and the Non-smokers Health Act and to make consequential amendments to other Acts 655 Third Avenue, 10th Floor, New York, NY 10017-5646, USA t: +1-212-642-1776 f: +1-212-768-7796 inta.org esanzdeacedo@inta.org The Honorable Kelvin Kenneth Ogilvie Chair Standing Committee on Social Affairs,

More information

PharmaSUG Paper QT38

PharmaSUG Paper QT38 PharmaSUG 2015 - Paper QT38 Statistical Review and Validation of Patient Narratives Indrani Sarkar, inventiv Health Clinical, Indiana, USA Bradford J. Danner, Sarah Cannon Research Institute, Tennessee,

More information